Empirical issues have played a key role in the recent literature on economic growth. Theoretical work on endogenous growth (for example Romer (1986) and Lucas (1988)) has often been motivated by the apparent inability of earlier neo-classical models to explain some important features of cross-country income and growth data. This research, in turn, has spurred a fast-growing empirical literature. In Discussion Paper No. 1275, Research Affiliate Angel de la Fuente makes a selective survey of this empirical branch of the growth literature, with special attention to its theoretical background and its implications for the evolution of the international distribution of income.

Recent empirical studies tend to confirm the importance of investment in human and technological capital as determinants of growth. Extensions of the neo-classical model that incorporate these factors explain rather well the long-run growth experience of a large sample of countries, and are consistent with the observed level of international inequality and its upward trend. These models do not explain very well the variation of growth rates over time, however, and the lack of robustness of the estimated coefficients to sample selection, specification, and other factors suggests that we are still far from having precise estimates of structural parameters such as the coefficients of the aggregate production function. Lastly, the methodology employed in these studies does not allow us to reach clear conclusions regarding the likely evolution of the international income distribution of income, or the policies which may have an impact on growth.